Will Computer-Aided Detection and Diagnosis Revolutionize Colonoscopy?

Artificial intelligence for colorectal neoplasia detection during colonoscopy: a systematic review and meta-analysis of randomized clinical trials

BACKGROUND

The use of artificial intelligence (AI) in detecting colorectal neoplasia during colonoscopy holds the potential to enhance adenoma detection rates (ADRs) and reduce adenoma miss rates (AMRs). However, varied outcomes have been observed across studies. Thus, this study aimed to evaluate the potential advantages and disadvantages of employing AI-aided systems during colonoscopy.

METHODS

Using Medical Subject Headings (MeSH) terms and keywords, a comprehensive electronic literature search was performed of the Embase, Medline, and the Cochrane Library databases from the inception of each database until October 04, 2023, in order to identify randomized controlled trials (RCTs) comparing AI-assisted with standard colonoscopy for detecting colorectal neoplasia. Primary outcomes included AMR, ADR, and adenomas detected per colonoscopy (APC). Secondary outcomes comprised the poly missed detection rate (PMR), poly detection rate (PDR), and poly detected per colonoscopy (PPC). We utilized random-effects meta-analyses with Hartung-Knapp adjustment to consolidate results. The prediction interval (PI) and I2 statistics were utilized to quantify between-study heterogeneity. Moreover, meta-regression and subgroup analyses were performed to investigate the potential sources of heterogeneity. This systematic review and meta-analysis is registered with PROSPERO (CRD42023428658).

FINDINGS

This study encompassed 33 trials involving 27,404 patients. Those undergoing AI-aided colonoscopy experienced a significant decrease in PMR (RR, 0.475; 95% CI, 0.294-0.768; I2 = 87.49%) and AMR (RR, 0.495; 95% CI, 0.390-0.627; I2 = 48.76%). Additionally, a significant increase in PDR (RR, 1.238; 95% CI, 1.158-1.323; I2 = 81.67%) and ADR (RR, 1.242; 95% CI, 1.159-1.332; I2 = 78.87%), along with a significant increase in the rates of PPC (IRR, 1.388; 95% CI, 1.270-1.517; I2 = 91.99%) and APC (IRR, 1.390; 95% CI, 1.277-1.513; I2 = 86.24%), was observed. This resulted in 0.271 more PPCs (95% CI, 0.144-0.259; I2 = 65.61%) and 0.202 more APCs (95% CI, 0.144-0.259; I2 = 68.15%).

INTERPRETATION

AI-aided colonoscopy significantly enhanced the detection of colorectal neoplasia detection, likely by reducing the miss rate. However, future studies should focus on evaluating the cost-effectiveness and long-term benefits of AI-aided colonoscopy in reducing cancer incidence.

FUNDING

This work was supported by the Heilongjiang Provincial Natural Science Foundation of China (LH2023H096), the Postdoctoral research project in Heilongjiang Province (LBH-Z22210), the National Natural Science Foundation of China's General Program (82072640) and the Outstanding Youth Project of Heilongjiang Natural Science Foundation (YQ2021H023).

Artificial intelligence in endoscopy: Overview, applications, and future directions

Since the emergence of artificial intelligence (AI) in medicine, endoscopy applications in gastroenterology have been at the forefront of innovations. The ever-increasing number of studies necessitates the need to organize and classify applications in a useful way. Separating AI capabilities by computer aided detection (CADe), diagnosis (CADx), and quality assessment (CADq) allows for a systematic evaluation of each application. CADe studies have shown promise in accurate detection of esophageal, gastric and colonic neoplasia as well as identifying sources of bleeding and Crohn's disease in the small bowel. While more advanced CADx applications employ optical biopsies to give further information to characterize neoplasia and grade inflammatory disease, diverse CADq applications ensure quality and increase the efficiency of procedures. Future applications show promise in advanced therapeutic modalities and integrated systems that provide multimodal capabilities. AI is set to revolutionize clinical decision making and performance of endoscopy.

Automated Detection of Anatomical Landmarks During Colonoscopy Using a Deep Learning Model

Background and aims

Identification and photo-documentation of the ileocecal valve (ICV) and appendiceal orifice (AO) confirm completeness of colonoscopy examinations. We aimed to develop and test a deep convolutional neural network (DCNN) model that can automatically identify ICV and AO, and differentiate these landmarks from normal mucosa and colorectal polyps.

Methods

We prospectively collected annotated full-length colonoscopy videos of 318 patients undergoing outpatient colonoscopies. We created three nonoverlapping training, validation, and test data sets with 25,444 unaltered frames extracted from the colonoscopy videos showing four landmarks/image classes (AO, ICV, normal mucosa, and polyps). A DCNN classification model was developed, validated, and tested in separate data sets of images containing the four different landmarks.

Results

After training and validation, the DCNN model could identify both AO and ICV in 18 out of 21 patients (85.7%). The accuracy of the model for differentiating AO from normal mucosa, and ICV from normal mucosa were 86.4% (95% CI 84.1% to 88.5%), and 86.4% (95% CI 84.1% to 88.6%), respectively. Furthermore, the accuracy of the model for differentiating polyps from normal mucosa was 88.6% (95% CI 86.6% to 90.3%).

Conclusion

This model offers a novel tool to assist endoscopists with automated identification of AO and ICV during colonoscopy. The model can reliably distinguish these anatomical landmarks from normal mucosa and colorectal polyps. It can be implemented into automated colonoscopy report generation, photo-documentation, and quality auditing solutions to improve colonoscopy reporting quality.

Effectiveness and application of artificial intelligence for endoscopic screening of colorectal cancer: the future is now

INTRODUCTION

Artificial intelligence (AI) in gastrointestinal endoscopy includes systems designed to interpret medical images and increase sensitivity during examination. This may be a promising solution to human biases and may provide support during diagnostic endoscopy.

AREAS COVERED

This review aims to summarize and evaluate data supporting AI technologies in lower endoscopy, addressing their effectiveness, limitations, and future perspectives.

EXPERT OPINION

Computer-aided detection (CADe) systems have been studied with promising results, allowing for an increase in adenoma detection rate (ADR), adenoma per colonoscopy (APC), and a reduction in adenoma miss rate (AMR). This may lead to an increase in the sensitivity of endoscopic examinations and a reduction in the risk of interval-colorectal cancer. In addition, computer-aided characterization (CADx) has also been implemented, aiming to distinguish adenomatous and non-adenomatous lesions through real-time assessment using advanced endoscopic imaging techniques. Moreover, computer-aided quality (CADq) systems have been developed with the aim of standardizing quality measures in colonoscopy (e.g. withdrawal time and adequacy of bowel cleansing) both to improve the quality of examinations and set a reference standard for randomized controlled trials.

The application of artificial intelligence in improving colonoscopic adenoma detection rate: Where are we and where are we going

Colorectal cancer (CRC) is one of the common malignant tumors in the world. Colonoscopy is the crucial examination technique in CRC screening programs for the early detection of precursor lesions, and treatment of early colorectal cancer, which can reduce the morbidity and mortality of CRC significantly. However, pooled polyp miss rates during colonoscopic examination are as high as 22%. Artificial intelligence (AI) provides a promising way to improve the colonoscopic adenoma detection rate (ADR). It might assist endoscopists in avoiding missing polyps and offer an accurate optical diagnosis of suspected lesions. Herein, we described some of the milestone studies in using AI for colonoscopy, and the future application directions of AI in improving colonoscopic ADR.

From advanced diagnosis to advanced resection in early neoplastic colorectal lesions: Never-ending and trending topics in the 2020s

Colonoscopy represents the most widespread and effective tool for the prevention and treatment of early stage preneoplastic and neoplastic lesions in the panorama of cancer screening. In the world there are different approaches to the topic of colorectal cancer prevention and screening: different starting ages (45-50 years); different initial screening tools such as fecal occult blood with immunohistochemical or immune-enzymatic tests; recto-sigmoidoscopy; and colonoscopy. The key aspects of this scenario are composed of a proper bowel preparation that ensures a valid diagnostic examination, experienced endoscopist in detection of preneoplastic and early neoplastic lesions and open-minded to upcoming artificial intelligence-aided examination, knowledge in the field of resection of these lesions (from cold-snaring, through endoscopic mucosal resection and endoscopic submucosal dissection, up to advanced tools), and management of complications.

Polyp detection in video colonoscopy using deep learning

Video colonoscopy automatic processing is a challenge and further development of computer assisted diagnosis is very helpful in correctness assessment of the exam, in e-learning and training, for statistics on polyps’ malignity or in polyps’ survey. New devices and programming languages are emerging and deep learning begun already to furnish astonishing results, in the quest for high speed and optimal polyp detection software. This paper presents a successful attempt in detecting the intestinal polyps in real time video colonoscopy with deep learning, using Mobile Net.

The location-based resect and discard strategy for diminutive colorectal polyps: a prospective clinical study

BACKGROUND

Clinical implementation of the resect-and-discard strategy has been difficult because optical diagnosis is highly operator dependent. This prospective study aimed to evaluate a resect-and-discard strategy that is not operator dependent.

METHODS

The study evaluated a resect-and-discard strategy that uses the anatomical polyp location to classify colonic polyps into non-neoplastic or low risk neoplastic. All rectosigmoid diminutive polyps were considered hyperplastic and all polyps located proximally to the sigmoid colon were considered neoplastic. Surveillance interval assignments based on these a priori assumptions were compared with those based on actual pathology results and on optical diagnosis. The primary outcome was ≥ 90 % agreement with pathology in surveillance interval assignment.

RESULTS

1117 patients undergoing complete colonoscopy were included and 482 (43.1 %) had at least one diminutive polyp. Surveillance interval agreement between the location-based strategy and pathological findings using the 2020 US Multi-Society Task Force guideline was 97.0 % (95 % confidence interval [CI] 0.96-0.98), surpassing the ≥ 90 % benchmark. Optical diagnoses using the NICE and Sano classifications reached 89.1 % and 90.01 % agreement, respectively (P < 0.001), and were inferior to the location-based strategy. The location-based resect-and-discard strategy allowed a 69.7 % (95 %CI 0.67-0.72) reduction in pathology examinations compared with 55.3 % (95 %CI 0.52-0.58; NICE and Sano) and 41.9 % (95 %CI 0.39-0.45; WASP) with optical diagnosis.

CONCLUSION

The location-based resect-and-discard strategy achieved very high surveillance interval agreement with pathology-based surveillance interval assignment, surpassing the ≥ 90 % benchmark and outperforming optical diagnosis in surveillance interval agreement and the number of pathology examinations avoided.

Initial application of deep learning to borescope detection of endoscope working channel damage and residue

Background and study aims  Outbreaks of endoscopy-related infections have prompted evaluation for potential contributing factors. We and others have demonstrated the utility of borescope inspection of endoscope working channels to identify occult damage that may impact the adequacy of endoscope reprocessing. The time investment and training necessary for borescope inspection have been cited as barriers preventing implementation. We investigated the utility of artificial intelligence (AI) for streamlining and enhancing the value of borescope inspection of endoscope working channels. Methods  We applied a deep learning AI approach to borescope inspection videos of the working channels of 20 endoscopes in use at our academic institution. We evaluated the sensitivity, accuracy, and reliability of this software for detection of endoscope working channel findings. Results  Overall sensitivity for AI-based detection of borescope inspection findings identified by gold standard endoscopist inspection was 91.4 %. Labels were accurate for 67 % of these working channel findings and accuracy varied by endoscope segment. Read-to-read variability was noted to be minimal, with test-retest correlation value of 0.986. Endoscope type did not predict accuracy of the AI system ( P  = 0.26). Conclusions  Harnessing the power of AI for detection of endoscope working channel damage and residue could enable sterile processing department technicians to feasibly assess endoscopes for working channel damage and perform endoscope reprocessing surveillance. Endoscopes that accumulate an unacceptable level of damage may be flagged for further manual evaluation and consideration for manufacturer evaluation/repair.

Artificial Intelligence-Aid Colonoscopy Vs. Conventional Colonoscopy for Polyp and Adenoma Detection: A Systematic Review of 7 Discordant Meta-Analyses

Objectives: Multiple meta-analyses which investigated the comparative efficacy and safety of artificial intelligence (AI)-aid colonoscopy (AIC) vs. conventional colonoscopy (CC) in the detection of polyp and adenoma have been published. However, a definitive conclusion has not yet been generated. This systematic review selected from discordant meta-analyses to draw a definitive conclusion about whether AIC is better than CC for the detection of polyp and adenoma. Methods: We comprehensively searched potentially eligible literature in PubMed, Embase, Cochrane library, and China National Knowledgement Infrastructure (CNKI) databases from their inceptions until to April 2021. Assessment of Multiple Systematic Reviews (AMSTAR) instrument was used to assess the methodological quality. Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) checklist was used to assess the reporting quality. Two investigators independently used the Jadad decision algorithm to select high-quality meta-analyses which summarized the best available evidence. Results: Seven meta-analyses met our selection criteria finally. AMSTAR score ranged from 8 to 10, and PRISMA score ranged from 23 to 26. According to the Jadad decision algorithm, two high-quality meta-analyses were selected. These two meta-analyses suggested that AIC was superior to CC for colonoscopy outcomes, especially for polyp detection rate (PDR) and adenoma detection rate (ADR). Conclusion: Based on the best available evidence, we conclude that AIC should be preferentially selected for the route screening of colorectal lesions because it has potential value of increasing the polyp and adenoma detection. However, the continued improvement of AIC in differentiating the shape and pathology of colorectal lesions is needed.

Artificial intelligence and computer-aided diagnosis for colonoscopy: where do we stand now?

Computer-aided diagnosis (CAD) for colonoscopy with use of artificial intelligence (AI) is catching increased attention of endoscopists. CAD allows automated detection and pathological prediction, namely optical biopsy, of colorectal polyps during real-time endoscopy, which help endoscopists avoid missing and/or misdiagnosing colorectal lesions. With the increased number of publications in this field and emergence of the AI medical device that have already secured regulatory approval, CAD in colonoscopy is now being implemented into clinical practice. On the other side, drawbacks and weak points of CAD in colonoscopy have not been thoroughly discussed. In this review, we provide an overview of CAD for optical biopsy of colorectal lesions with a particular focus on its clinical applications and limitations.

Weakly Supervised Attention Map Training for Histological Localization of Colonoscopy Images

We consider the problem of training a convolutional neural network for histological localization of colorectal lesions from imperfectly annotated datasets. Given that we have a colonoscopic image dataset for 4-class histology classification and another dataset originally dedicated to polyp segmentation, we propose a weakly supervised learning approach to histological localization by training with the two different types of datasets. With the classification dataset, we first train a convolutional neural network to classify colonoscopic images into 4 different histology categories. By interpreting the trained classifier, we can extract an attention map corresponding to the predicted class for each colonoscopy image. We further improve the localization accuracy of attention maps by training the model to focus on lesions under the guidance of the polyp segmentation dataset. The experimental results show that the proposed approach simultaneously improves histology classification and lesion localization accuracy.

Artificial intelligence and colonoscopy − enhancements and improvements

Artificial intelligence is a technology that processes and analyzes information with reproducibility and accuracy. Its application in medicine, especially in the field of gastroenterology, has great potential to facilitate in diagnosis of various disease states. Currently, the role of artificial intelligence as it pertains to colonoscopy revolves around enhanced polyp detection and characterization. The aim of this article is to review the current and potential future applications of artificial intelligence for enhanced quality of detection for colorectal neoplasia.

Artificial intelligence can increase the detection rate of colorectal polyps and adenomas: a systematic review and meta-analysis

Colonoscopy is an important method to diagnose polyps, especially adenomatous polyps. However, the rate of missed diagnoses is relatively high. In this study, we aimed to determine whether artificial intelligence (AI) improves the polyp detection rate (PDR) and adenoma detection rate (ADR) with colonoscopy. We performed a systematic search in PubMed, Cochrane Library, Embase, and Web of Science databases; the search included entries in the databases up to and including 29 February 2020. Five articles that involved a total of 4311 patients fulfilled the selection criteria. The results of these studies showed that both PDR and ADR increased with the assistance of AI compared with those in control groups {pooled odds ratio (OR) = 1.91 [95% confidence interval (CI) 1.68-2.16] and 1.75 (95% CI 1.52-2.01), respectively}. Good bowel preparation reduced the impact of AI, but significant differences were still apparent in PDR and ADR [pooled OR = 1.69 (95% CI 1.32-2.16) and 1.36 (95% CI 1.04-1.78), respectively]. The characteristics of polyps and adenomas also influenced the results. The average number of polyps and adenomas detected varied significantly by location, and small polyps and adenomas were more likely to be missed. However, the effect of the morphology of polyps and AI-assisted detection needs further studies. In conclusion, AI increases the detection rates of polyps and adenomas in colonoscopy. Without AI assistance, detection rates can be improved with better bowel preparation and training for small polyp and adenoma detection.

Use of Artificial Intelligence to Improve the Quality Control of Gastrointestinal Endoscopy

With the rapid development of science and technology, artificial intelligence (AI) systems are becoming ubiquitous, and their utility in gastroenteroscopy is beginning to be recognized. Digestive endoscopy is a conventional and reliable method of examining and diagnosing digestive tract diseases. However, with the increase in the number and types of endoscopy, problems such as a lack of skilled endoscopists and difference in the professional skill of doctors with different degrees of experience have become increasingly apparent. Most studies thus far have focused on using computers to detect and diagnose lesions, but improving the quality of endoscopic examination process itself is the basis for improving the detection rate and correctly diagnosing diseases. In the present study, we mainly reviewed the role of AI in monitoring systems, mainly through the endoscopic examination time, reducing the blind spot rate, improving the success rate for detecting high-risk lesions, evaluating intestinal preparation, increasing the detection rate of polyps, automatically collecting maps and writing reports. AI can even perform quality control evaluations for endoscopists, improve the detection rate of endoscopic lesions and reduce the burden on endoscopists.

Prediction of the histology of colorectal neoplasm in white light colonoscopic images using deep learning algorithms

The treatment plan of colorectal neoplasm differs based on histology. Although new endoscopic imaging systems have been developed, there are clear diagnostic thresholds and requirements in using them. To overcome these limitations, we trained convolutional neural networks (CNNs) with endoscopic images and developed a computer-aided diagnostic (CAD) system which predicts the pathologic histology of colorectal adenoma. We retrospectively collected colonoscopic images from two tertiary hospitals and labeled 3400 images into one of 4 classes according to the final histology: normal, low-grade dysplasia, high-grade dysplasia, and adenocarcinoma. We implemented a CAD system based on ensemble learning with three CNN models which transfer the knowledge learned from common digital photography images to the colonoscopic image domain. The deep learning models were trained to classify the colorectal adenoma into these 4 classes. We compared the outcomes of the CNN models to those of two endoscopist groups having different years of experience, and visualized the model predictions using Class Activation Mapping. In our multi-center study, our CNN-CAD system identified the histology of colorectal adenoma with as sensitivity 77.25%, specificity of 92.42%, positive predictive value of 77.16%, negative predictive value of 92.58% averaged over the 4 classes, and mean diagnostic time of 0.12 s per image. Our experiments demonstrate that the CNN-CAD showed a similar performance to that of endoscopic experts and outperformed that of trainees. The model visualization results also showed reasonable regions of interest to explain the classification decisions of CAD systems. We suggest that CNN-CAD system can predict the histology of colorectal adenoma.

Application of Artificial Intelligence in Gastrointestinal Endoscopy

Artificial intelligence (AI), also known as computer-aided diagnosis, is a technology that enables machines to process information and functions at or above human level and has great potential in gastrointestinal endoscopy applications. At present, the research on medical image recognition usually adopts the deep-learning algorithm based on the convolutional neural network. AI has been used in gastrointestinal endoscopy including esophagogastroduodenoscopy, capsule endoscopy, colonoscopy, etc. AI can help endoscopic physicians improve the diagnosis rate of various lesions, reduce the rate of missed diagnosis, improve the quality of endoscopy, assess the severity of the disease, and improve the efficiency of endoscopy. The diversity, susceptibility, and imaging specificity of gastrointestinal endoscopic images are all difficulties and challenges on the road to intelligence. We need more large-scale, high-quality, multicenter prospective studies to explore the clinical applicability of AI, and ethical issues need to be taken into account.

Artificial Intelligence-Assisted Colonoscopy for Detection of Colon Polyps: a Prospective, Randomized Cohort Study

BACKGROUND AND AIMS

Improving the rate of polyp detection is an important measure to prevent colorectal cancer (CRC). Real-time automatic polyp detection systems, through deep learning methods, can learn and perform specific endoscopic tasks previously performed by endoscopists. The purpose of this study was to explore whether a high-performance, real-time automatic polyp detection system could improve the polyp detection rate (PDR) in the actual clinical environment.

METHODS

The selected patients underwent same-day, back-to-back colonoscopies in a random order, with either traditional colonoscopy or artificial intelligence (AI)-assisted colonoscopy performed first by different experienced endoscopists (> 3000 colonoscopies). The primary outcome was the PDR. It was registered with clinicaltrials.gov . (NCT047126265).

RESULTS

In this study, we randomized 150 patients. The AI system significantly increased the PDR (34.0% vs 38.7%, p < 0.001). In addition, AI-assisted colonoscopy increased the detection of polyps smaller than 6 mm (69 vs 91, p < 0.001), but no difference was found with regard to larger lesions.

CONCLUSIONS

A real-time automatic polyp detection system can increase the PDR, primarily for diminutive polyps. However, a larger sample size is still needed in the follow-up study to further verify this conclusion.

TRIAL REGISTRATION

clinicaltrials.gov Identifier: NCT047126265.

PICCOLO White-Light and Narrow-Band Imaging Colonoscopic Dataset: A Performance Comparative of Models and Datasets

Colorectal cancer is one of the world leading death causes. Fortunately, an early diagnosis allows for effective treatment, increasing the survival rate. Deep learning techniques have shown their utility for increasing the adenoma detection rate at colonoscopy, but a dataset is usually required so the model can automatically learn features that characterize the polyps. In this work, we present the PICCOLO dataset, that comprises 3433 manually annotated images (2131 white-light images 1302 narrow-band images), originated from 76 lesions from 40 patients, which are distributed into training (2203), validation (897) and test (333) sets assuring patient independence between sets. Furthermore, clinical metadata are also provided for each lesion. Four different models, obtained by combining two backbones and two encoder–decoder architectures, are trained with the PICCOLO dataset and other two publicly available datasets for comparison. Results are provided for the test set of each dataset. Models trained with the PICCOLO dataset have a better generalization capacity, as they perform more uniformly along test sets of all datasets, rather than obtaining the best results for its own test set. This dataset is available at the website of the Basque Biobank, so it is expected that it will contribute to the further development of deep learning methods for polyp detection, localisation and classification, which would eventually result in a better and earlier diagnosis of colorectal cancer, hence improving patient outcomes.

Use of artificial intelligence in improving adenoma detection rate during colonoscopy: Might both endoscopists and pathologists be further helped

Colonoscopy remains the standard strategy for screening for colorectal cancer around the world due to its efficacy in both detecting adenomatous or pre-cancerous lesions and the capacity to remove them intra-procedurally. Computer-aided detection and diagnosis (CAD), thanks to the brand new developed innovations of artificial intelligence, and especially deep-learning techniques, leads to a promising solution to human biases in performance by guarantying decision support during colonoscopy. The application of CAD on real-time colonoscopy helps increasing the adenoma detection rate, and therefore contributes to reduce the incidence of interval cancers improving the effectiveness of colonoscopy screening on critical outcome such as colorectal cancer related mortality. Furthermore, a significant reduction in costs is also expected. In addition, the assistance of the machine will lead to a reduction of the examination time and therefore an optimization of the endoscopic schedule. The aim of this opinion review is to analyze the clinical applications of CAD and artificial intelligence in colonoscopy, as it is reported in literature, addressing evidence, limitations, and future prospects.

Clinical Trials for Artificial Intelligence in Cancer Diagnosis: A Cross-Sectional Study of Registered Trials in ClinicalTrials.gov

Objective: Clinical trials are the most effective way to judge the merits of diagnosis and treatment strategies. The in-depth mining of clinical trial data enables us to grasp the application trend of artificial intelligence (AI) for cancer diagnosis. The aim of this study was to analyze the characteristics of registered trials on AI for cancer diagnosis. Methods: Clinical trials on AI for cancer diagnosis registered on the ClinicalTrials.gov database were searched and downloaded. Statistical analysis was performed by using SPSS 20.0 software. Results: A total of 97 registered trials were included. Of them, only 27 (27.8%) were interventional trials and 70 (72.1%) were observational trials. Fifteen (15.4%) trials had been completed. Fifty trials were in recruitment, and another 18 remained unrecruited. The number of cases included in the clinical trials tended to be large, 31 (32.0%) trials including samples ranging from 100 to 499 cases and 17 (17.5%) trials including samples ranging from 500 to 999 cases. Of the 27 interventional trials, only two trials reported trials' phase. Most (85.2%) interventional trials were for diagnosis, and a few (3.7%) were for the purpose of both the diagnosis and therapy of cancers. For the observational clinical trials, 46 (65.7%) were cohort studies, and 11 (15.7%) were case-only studies. Among the observational trials, 46 (65.7%) were prospective studies and 13 (18.6%) were retrospective studies. Among 97 trials, 37 (38.1%) involved colorectal cancer, 11 (11.3%) involved breast cancer, 43 (44.3%) were for imaging diagnosis, 33 (34.0%) were for endoscopic diagnosis, and 11 (11.3%) were for pathological diagnosis. For the interventional trials, 11 trials were parallel assignment (40.7%), and 14 were single group assignment (51.9%). Among the 27 interventional trials, 18 (66.7%) trials were performed without masking, 6 (22.2%) trials were performed with single masking, only 1 (3.7%) was performed with double masking, and 2 (7.4%) was performed with triple masking. Conclusion: It appears that most registered trials on AI for cancer diagnosis are observational design, and more trials are needed in this field.

Eigenloss: Combined PCA-Based Loss Function for Polyp Segmentation

Colorectal cancer is one of the leading cancer death causes worldwide, but its early diagnosis highly improves the survival rates. The success of deep learning has also benefited this clinical field. When training a deep learning model, it is optimized based on the selected loss function. In this work, we consider two networks (U-Net and LinkNet) and two backbones (VGG-16 and Densnet121). We analyzed the influence of seven loss functions and used a principal component analysis (PCA) to determine whether the PCA-based decomposition allows for the defining of the coefficients of a non-redundant primal loss function that can outperform the individual loss functions and different linear combinations. The eigenloss is defined as a linear combination of the individual losses using the elements of the eigenvector as coefficients. Empirical results show that the proposed eigenloss improves the general performance of individual loss functions and outperforms other linear combinations when Linknet is used, showing potential for its application in polyp segmentation problems.

Deep learning to find colorectal polyps in colonoscopy: A systematic literature review

Colorectal cancer has a great incidence rate worldwide, but its early detection significantly increases the survival rate. Colonoscopy is the gold standard procedure for diagnosis and removal of colorectal lesions with potential to evolve into cancer and computer-aided detection systems can help gastroenterologists to increase the adenoma detection rate, one of the main indicators for colonoscopy quality and predictor for colorectal cancer prevention. The recent success of deep learning approaches in computer vision has also reached this field and has boosted the number of proposed methods for polyp detection, localization and segmentation. Through a systematic search, 35 works have been retrieved. The current systematic review provides an analysis of these methods, stating advantages and disadvantages for the different categories used; comments seven publicly available datasets of colonoscopy images; analyses the metrics used for reporting and identifies future challenges and recommendations. Convolutional neural networks are the most used architecture together with an important presence of data augmentation strategies, mainly based on image transformations and the use of patches. End-to-end methods are preferred over hybrid methods, with a rising tendency. As for detection and localization tasks, the most used metric for reporting is the recall, while Intersection over Union is highly used in segmentation. One of the major concerns is the difficulty for a fair comparison and reproducibility of methods. Even despite the organization of challenges, there is still a need for a common validation framework based on a large, annotated and publicly available database, which also includes the most convenient metrics to report results. Finally, it is also important to highlight that efforts should be focused in the future on proving the clinical value of the deep learning based methods, by increasing the adenoma detection rate.

Computer-Aided Diagonosis for Colorectal Cancer using Deep Learning with Visual Explanations

Detection, diagnosis, and removal of colorectal neoplasms are well-accepted colorectal cancer prevention methods. Although promising endoscopic imaging techniques including narrow-band imaging have been developed, these techniques are operator-dependent and interpretations of the results may vary. To overcome these limitations, we applied deep learning to develop a computer-aided diagnostic (CAD) system of colorectal adenoma. We collected and divided 3000 colonoscopic images into 4 categories according to the final pathology, normal, low-grade dysplasia, high-grade dysplasia, and adenocarcinoma. We implemented three convolutional neural networks (CNNs) using Inception-v3, ResNet-50, and DenseNet-161 as baseline models. We further altered the models using several strategies: replacement of the top layer, transfer learning from pre-trained models, fine-tuning of the model weights, rebalancing and augmentation of the training data, and 10-fold cross-validation. We compared the outcomes of the three CNN models to those of two endoscopist groups having different years of experience, and visualized the model predictions using Class Activation Mapping (CAM). The CNN-CAD achieved the best performance in our experiments with a 92.48% classification accuracy rate. The CNN-CAD results showed a better performance in all criteria than those of endoscopic experts. The model visualization results showed reasonable regions of interest to explain pathology classification decisions. We demonstrated that CNN-CAD can distinguish the pathology of colorectal adenoma, yielding better outcomes than the endoscopic experts group.

Optical evaluation: the crux for effective management of colorectal neoplasia

Advances in minimally invasive tissue resection techniques now allow for the majority of early colorectal neoplasia to be managed endoscopically. To optimize their respective risk-benefit profiles, and, therefore, appropriately select between endoscopic mucosal resection, endoscopic submucosal dissection, and surgery, the endoscopist must accurately predict the risk of submucosal invasive cancer and estimate depth of invasion. Herein, we discuss the evidence and our approach for optical evaluation of large (⩾ 20 mm) colorectal laterally spreading lesions.

Automated endoscopic detection and classification of colorectal polyps using convolutional neural networks

BACKGROUND

Recently the American Society for Gastrointestinal Endoscopy addressed the 'resect and discard' strategy, determining that accurate in vivo differentiation of colorectal polyps (CP) is necessary. Previous studies have suggested a promising application of artificial intelligence (AI), using deep learning in object recognition. Therefore, we aimed to construct an AI system that can accurately detect and classify CP using stored still images during colonoscopy.

METHODS

We used a deep convolutional neural network (CNN) architecture called Single Shot MultiBox Detector. We trained the CNN using 16,418 images from 4752 CPs and 4013 images of normal colorectums, and subsequently validated the performance of the trained CNN in 7077 colonoscopy images, including 1172 CP images from 309 various types of CP. Diagnostic speed and yields for the detection and classification of CP were evaluated as a measure of performance of the trained CNN.

RESULTS

The processing time of the CNN was 20 ms per frame. The trained CNN detected 1246 CP with a sensitivity of 92% and a positive predictive value (PPV) of 86%. The sensitivity and PPV were 90% and 83%, respectively, for the white light images, and 97% and 98% for the narrow band images. Among the correctly detected polyps, 83% of the CP were accurately classified through images. Furthermore, 97% of adenomas were precisely identified under the white light imaging.

CONCLUSIONS

Our CNN showed promise in being able to detect and classify CP through endoscopic images, highlighting its high potential for future application as an AI-based CP diagnosis support system for colonoscopy.

CAD-CAP: a 25,000-image database serving the development of artificial intelligence for capsule endoscopy

Background and study aims  Capsule endoscopy (CE) is the preferred method for small bowel (SB) exploration. With a mean number of 50,000 SB frames per video, SBCE reading is time-consuming and tedious (30 to 60 minutes per video). We describe a large, multicenter database named CAD-CAP (Computer-Assisted Diagnosis for CAPsule Endoscopy, CAD-CAP). This database aims to serve the development of CAD tools for CE reading. Materials and methods  Twelve French endoscopy centers were involved. All available third-generation SB-CE videos (Pillcam, Medtronic) were retrospectively selected from these centers and deidentified. Any pathological frame was extracted and included in the database. Manual segmentation of findings within these frames was performed by two pre-med students trained and supervised by an expert reader. All frames were then classified by type and clinical relevance by a panel of three expert readers. An automated extraction process was also developed to create a dataset of normal, proofread, control images from normal, complete, SB-CE videos. Results  Four-thousand-one-hundred-and-seventy-four SB-CE were included. Of them, 1,480 videos (35 %) containing at least one pathological finding were selected. Findings from 5,184 frames (with their short video sequences) were extracted and delimited: 718 frames with fresh blood, 3,097 frames with vascular lesions, and 1,369 frames with inflammatory and ulcerative lesions. Twenty-thousand normal frames were extracted from 206 SB-CE normal videos. CAD-CAP has already been used for development of automated tools for angiectasia detection and also for two international challenges on medical computerized analysis.

A novel artificial intelligence system for the assessment of bowel preparation (with video)

BACKGROUND AND AIMS

The quality of bowel preparation is an important factor that can affect the effectiveness of a colonoscopy. Several tools, such as the Boston Bowel Preparation Scale (BBPS) and Ottawa Bowel Preparation Scale, have been developed to evaluate bowel preparation. However, understanding the differences between evaluation methods and consistently applying them can be challenging for endoscopists. There are also subjective biases and differences among endoscopists. Therefore, this study aimed to develop a novel, objective, and stable method for the assessment of bowel preparation through artificial intelligence.

METHODS

We used a deep convolutional neural network to develop this novel system. First, we retrospectively collected colonoscopy images to train the system and then compared its performance with endoscopists via a human-machine contest. Then, we applied this model to colonoscopy videos and developed a system named ENDOANGEL to provide bowel preparation scores every 30 seconds and to show the cumulative ratio of frames for each score during the withdrawal phase of the colonoscopy.

RESULTS

ENDOANGEL achieved 93.33% accuracy in the human-machine contest with 120 images, which was better than that of all endoscopists. Moreover, ENDOANGEL achieved 80.00% accuracy among 100 images with bubbles. In 20 colonoscopy videos, accuracy was 89.04%, and ENDOANGEL continuously showed the accumulated percentage of the images for different BBPS scores during the withdrawal phase and prompted us for bowel preparation scores every 30 seconds.

CONCLUSIONS

We provided a novel and more accurate evaluation method for bowel preparation and developed an objective and stable system-ENDOANGEL-that could be applied reliably and steadily in clinical settings.

Prediction of Polyp Pathology Using Convolutional Neural Networks Achieves "Resect and Discard" Thresholds

OBJECTIVES

Reliable in situ diagnosis of diminutive (≤5 mm) colorectal polyps could allow for "resect and discard" and "diagnose and leave" strategies, resulting in $1 billion cost savings per year in the United States alone. Current methodologies have failed to consistently meet the Preservation and Incorporation of Valuable endoscopic Innovations (PIVIs) initiative thresholds. Convolutional neural networks (CNNs) have the potential to predict polyp pathology and achieve PIVI thresholds in real time.

METHODS

We developed a CNN-based optical pathology (OP) model using Tensorflow and pretrained on ImageNet, capable of operating at 77 frames per second. A total of 6,223 images of unique colorectal polyps of known pathology, location, size, and light source (white light or narrow band imaging [NBI]) underwent 5-fold cross-training (80%) and validation (20%). Separate fresh validation was performed on 634 polyp images. Surveillance intervals were calculated, comparing OP with true pathology.

RESULTS

In the original validation set, the negative predictive value for adenomas was 97% among diminutive rectum/rectosigmoid polyps. Results were independent of use of NBI or white light. Surveillance interval concordance comparing OP and true pathology was 93%. In the fresh validation set, the negative predictive value was 97% among diminutive polyps in the rectum and rectosigmoid and surveillance concordance was 94%.

DISCUSSION

This study demonstrates the feasibility of in situ diagnosis of colorectal polyps using CNN. Our model exceeds PIVI thresholds for both "resect and discard" and "diagnose and leave" strategies independent of NBI use. Point-of-care adenoma detection rate and surveillance recommendations are potential added benefits.

Strategies for Colorectal Cancer Screening

The incidence of colorectal cancer (CRC) is increasing worldwide. CRC has high mortality when detected at advanced stages, yet it is also highly preventable. Given the difficulties in implementing major lifestyle changes or widespread primary prevention strategies to decrease CRC risk, screening is the most powerful public health tool to reduce mortality. Screening methods are effective but have limitations. Furthermore, many screen-eligible people remain unscreened. We discuss established and emerging screening methods, and potential strategies to address current limitations in CRC screening. A quantum step in CRC prevention might come with the development of new screening strategies, but great gains can be made by deploying the available CRC screening modalities in ways that optimize outcomes while making judicious use of resources.

Artificial Intelligence in Gastrointestinal Endoscopy in a Resource-constrained Setting: A Reality Check

Artificial intelligence (AI) is being increasingly explored in different domains of gastroenterology, particularly in endoscopic image analysis, cancer screening, and prognostication models. It is widely touted to become an integral part of routine endoscopies, considering the bulk of data handled by endoscopists and the complex nature of critical analyses performed. However, the application of AI in endoscopy in resource-constrained settings remains fraught with problems. We conducted an extensive literature review using the PubMed database on articles covering the application of AI in endoscopy and the difficulties encountered in resource-constrained settings. We have tried to summarize in the present review the potential problems that may hinder the application of AI in such settings. Hopefully, this review will enable endoscopists and health policymakers to ponder over these issues before trying to extrapolate the advancements of AI in technically advanced settings to those having constraints at multiple levels.

How to cite this article: Anirvan P, Meher D, Singh SP. Artificial Intelligence in Gastrointestinal Endoscopy in a Resource-constrained Setting: A Reality Check. Euroasian J Hepato-Gastroenterol 2020;10(2): 92–97.

Study on detection rate of polyps and adenomas in artificial-intelligence-aided colonoscopy

BACKGROUND/AIM

To study the impact of computer-aided detection (CADe) system on the detection rate of polyps and adenomas in colonoscopy.

MATERIALS AND METHODS

A total of 1026 patients were prospectively randomly scheduled for colonoscopy with (the CADe group, CADe) or without (the control group, CON) the aid of the CADe system, together with visual notification and voice alarm, so as to compare the detection rate of polyp.

RESULTS

Compared with group CON, the detection rate of adenomas increased in group CADe, the average number of adenomas increased, the number of small adenomas increased, the number of proliferative polyps increased, and the differences were statistically significant (P < 0.001), but the comparison for the number of larger adenomas showed no significant difference between the groups (P> 0.05).

CONCLUSIONS

The CADe system is feasible for increasing the detection of polyps and adenomas in colonoscopy.

Use of image-enhanced endoscopy in the characterization of colorectal polyps: Still some ways to go

BACKGROUND/AIM

Instrument-based image-enhanced endoscopy (IEE) is of benefit in detecting and characterizing lesions during colonoscopy. We aimed to study the ability of community-based gastroenterologists to differentiate between neoplastic and non-neoplastic lesions using IEE modalities and to identify predictors of correct classification and the confidence of the optical diagnosis made.

MATERIALS AND METHODS:

An electronic survey was sent to practicing gastroenterologists using electronic tablets during a gastroenterology meeting. Demographic and professional information was gathered and endoscopic images of various colonic lesions were shown and they were requested to classify the images based in white light, flexible spectral imaging color enhancement (FICE), iScan, and narrow band imaging (NBI).

RESULTS:

Overall, 71 gastroenterologists responded to the survey, 76% were males and the majority were aged between 36 and 45 years (44%). Most of the respondents practiced both hepatology and gastroenterology (56%) and most of them had never received any training on IEE (66%). Correct identification of lesions using regular white light endoscopy was low (range 28%-84%). None of the IEE modalities increased the percentage of correct diagnoses apart from one NBI image where it increased from 28% (95%CI: 17%-38%) to 56% (95%CI: 44%-68%) (P < 0.01). Those who identified themselves as practicing mainly luminal gastroenterology were more confident 72% (95%CI: 60%-84%) compared with hepatologists 36% (95%CI: 25%-48%), or those who practiced both 48% (95%CI: 39%-56%) despite no difference in the percentage in correct answers.

CONCLUSION:

There remain areas of improvement in the performance of endoscopists in practice and would recommend more dedicated training programs, which could make use of asynchronous technological platforms.

Artificial intelligence and computer-aided diagnosis in colonoscopy: current evidence and future directions

Computer-aided diagnosis offers a promising solution to reduce variation in colonoscopy performance. Pooled miss rates for polyps are as high as 22%, and associated interval colorectal cancers after colonoscopy are of concern. Optical biopsy, whereby in-vivo classification of polyps based on enhanced imaging replaces histopathology, has not been incorporated into routine practice because it is limited by interobserver variability and generally only meets accepted standards in expert settings. Real-time decision-support software has been developed to detect and characterise polyps, and also to offer feedback on the technical quality of inspection. Some of the current algorithms, particularly with recent advances in artificial intelligence techniques, match human expert performance for optical biopsy. In this Review, we summarise the evidence for clinical applications of computer-aided diagnosis and artificial intelligence in colonoscopy.

Artificial intelligence in gastrointestinal endoscopy: The future is almost here

Artificial intelligence (AI) enables machines to provide unparalleled value in a myriad of industries and applications. In recent years, researchers have harnessed artificial intelligence to analyze large-volume, unstructured medical data and perform clinical tasks, such as the identification of diabetic retinopathy or the diagnosis of cutaneous malignancies. Applications of artificial intelligence techniques, specifically machine learning and more recently deep learning, are beginning to emerge in gastrointestinal endoscopy. The most promising of these efforts have been in computer-aided detection and computer-aided diagnosis of colorectal polyps, with recent systems demonstrating high sensitivity and accuracy even when compared to expert human endoscopists. AI has also been utilized to identify gastrointestinal bleeding, to detect areas of inflammation, and even to diagnose certain gastrointestinal infections. Future work in the field should concentrate on creating seamless integration of AI systems with current endoscopy platforms and electronic medical records, developing training modules to teach clinicians how to use AI tools, and determining the best means for regulation and approval of new AI technology.

Deep learning and conditional random fields-based depth estimation and topographical reconstruction from conventional endoscopy

Colorectal cancer is the fourth leading cause of cancer deaths worldwide and the second leading cause in the United States. The risk of colorectal cancer can be mitigated by the identification and removal of premalignant lesions through optical colonoscopy. Unfortunately, conventional colonoscopy misses more than 20% of the polyps that should be removed, due in part to poor contrast of lesion topography. Imaging depth and tissue topography during a colonoscopy is difficult because of the size constraints of the endoscope and the deforming mucosa. Most existing methods make unrealistic assumptions which limits accuracy and sensitivity. In this paper, we present a method that avoids these restrictions, using a joint deep convolutional neural network-conditional random field (CNN-CRF) framework for monocular endoscopy depth estimation. Estimated depth is used to reconstruct the topography of the surface of the colon from a single image. We train the unary and pairwise potential functions of a CRF in a CNN on synthetic data, generated by developing an endoscope camera model and rendering over 200,000 images of an anatomically-realistic colon.We validate our approach with real endoscopy images from a porcine colon, transferred to a synthetic-like domain via adversarial training, with ground truth from registered computed tomography measurements. The CNN-CRF approach estimates depths with a relative error of 0.152 for synthetic endoscopy images and 0.242 for real endoscopy images. We show that the estimated depth maps can be used for reconstructing the topography of the mucosa from conventional colonoscopy images. This approach can easily be integrated into existing endoscopy systems and provides a foundation for improving computer-aided detection algorithms for detection, segmentation and classification of lesions.

Spatially resolved diffuse reflectance spectroscopy endoscopic sensing with custom Si photodetectors

Early detection and surveillance of disease progression in epithelial tissue is key to improving long term patient outcomes for colon and esophageal cancers, which account for nearly a quarter of cancer related mortalities worldwide. Spatially resolved diffuse reflectance spectroscopy (SRDRS) is a non-invasive optical technique to sense biological changes at the cellular and sub-cellular level that occur when normal tissue becomes diseased, and has the potential to significantly improve the current standard of care for endoscopic gastrointestinal (GI) screening. Herein the design, fabrication, and characterization of the first custom SRDRS device to enable endoscopic SRDRS GI tissue characterization using a custom silicon (Si) thin film multi-pixel endoscopic optical sensor (MEOS) is described.